Only system cookies are currently stored. Tracking of activities and analytics features are turned off.

At Jetski Army, we have exciting offers and news about products and services that we hope you'd like to hear about. To present you the offers that suit you the most, we need to know a few personal details about you. We will gather some of your activities on our website (such as which pages you've visited, etc.) and use them to personalize the website content and improve analytics about our visitors. In addition, we will store small piecies of data in your browser cookies. We promise we will treat your data with respect, store it in a secured storage, and won't release it to any third parties.

Your Ride

Tell us your ride so we can recommend articles and products.

Articles / Racing / Measuring Engine Reliability

Measuring Engine Reliability

17 September 2018
Article Measuring Engine Reliability
First, of course, to gain more power from the new larger engine…..But of equal importance is to make this added power with a bigger margin of long term reliability.  Even though these new larger motors make better power, most owners still feel the need to do additional mods to get more power yet.  This begs the question, how do you “measure” the reliability of an engine setup??.... there is a way.

The most popular means for race engine builders and engineers to compare the reliability of different engine platforms is via “average piston speed”.  Knowing the average piston speed gives an indicator of how often you may need to replace internal moving parts (rods,bearings,pistons,etc)

So the million dollar question is “how high can piston-speed be with good reliability”.  Sadly there is no definitive number, but there are some very good parameters.  Race engine builders don’t obsess over finding a “safe” piston speed.  Instead, what engine builders really want to know is, “at what piston speed can I predict the eminent failure of an internal moving part…. So I can change that part before it breaks”.

 “In general”, it is not possible to predict the eminent failure of internal engine parts whose average piston speed exceeds 4000 feet per minute.  What this means is that at 4000 fpm, a connecting rod may fail after full load operation of 40 minutes, or 40 seconds.  In general, the farther you get away from 4000 fpm, the better (and more predictable) long term reliability gets.  Calculating the average piston speed is pretty simple math … See below.
Stroke Engine 7000rpm 7300rpm 7600rpm 4000fpm
60.0m Kaw 400-440-550 2760fpm 2870fpm 2990fpm 10,020r
68.0m Yam 633 701 760 
Yam1100/1200 un-valved
Sea Doo 720  all
3120fpm 3260fpm 3390fpm 8970rpm
69.2m Kaw SXR 1500 3170fpm 3310fpm 3440fpm 8820rpm
70.0m Kaw 650 all 3210fpm 3350fpm 3490fpm 8700rpm
71.0m Kaw 900 1100 triple 3260fpm 3400fpm 3540ppm 8590rpm
74.0m Kaw 750 800 all years
Yam 6mm stroker twins
Sea Doo 785  all
3400fpm 3540fpm 3690fpm 8240rpm
78.0m Yam GP800  GPR1200 3580fpm 3740fpm 3890fpm 7810rpm
78.2m Sea Doo 951  all 3590fpm 3745fpm 3900fpm 7800rpm
79.0m Kaw KX500  Honda CR500 3630fpm 3780fpm 3940fpm 7720rpm
Understanding the stresses of “a revolution”
While the piston speed data seems very straight forward, there is a lot more to it than that.  4000 fpm calculates out to about 45mph .... and that number seems pretty safe …. But it doesn’t work that way.  In truth, a piston in an SXR800 turning 4000fpm (8200rpm) starts a revolution from a dead stop at TDC.  From that stop, it accelerates to about 120+mph in 1.45 inches (where the crankpin is at 90’ from TDC).  From that 90’ position, the piston then decelerates from 120mph to 0 in the next 1.45 inches.  The sheer forces of these non-stop accelerations and decelerations is what breaks rods, pistons, bearings, etc.  The good news is that there are three ways to reduce the leathal forces of these accelerations/decelerations.  The first of course, is to reduce peak rpm numbers (which reduces the intensity of the accelerations/decelerations).  The second is to reduce the amount of time that the engine spends at these peak rpms (so the rod is subjected to fewer peak-rpm acceleration/deceleration cycles) … and lastly, install a longer connecting rod.  The geometry of a longer rod actually reduces the rate of acceleration and deceleration because it reduces the rod angle at 90’ from tdc and bdc.

In the case of high performance pwc engines, most of them operate at rpms very close to peak for a huge percentile of operating time (compared to an MX bike)…. So relying on staying away from peak rpm is not a wise option.  The next choice is to reduce the sheer peak rpm (average piston speed).  The truth is that good long term reliability (in a pwc application) can be had at piston speeds around 3300-3400 fpm.  Having a particularly robust crankshaft and long connecting rods can raise that range a bit…. But not by much.  The Sea Doo 951s and Yamaha GPR 800/1200 have very robust crankshafts, but that only buys them safety margins up to 3600fpm….. After that, the risk of connecting rod failure (breaking) becomes very high.
So what about longer Connecting Rods. –  In truth, it’s not just rod length that’s important … it’s the rod length to stroke ratio. In the early 1980s, Yamaha released its new YZ465 motocrosser.  The 465 shared many lower end bearing dimensions with the 250cc version…. But the 250 had a 68mm stroke vs. the 82mm stroke of the 465.  The rod bearing diameters of the two bikes was the same, but the 465 had a slightly longer connecting rod.  It wasn’t long before mechanics were fitting the longer 465 rods in to the 250 motors.  The installation was easy, and only required the addition of a spacer plate under the cylinder.  The performance results were noticeable, and all positive.  The apparent reduction in rod angle at 90’btc/atc resulted in broader torque curve, and a more rideable power delivery.  In addition, the long term wear of the piston and cylinder wall were considerably better.  In time, future YZ250s were fitted with the same longer rod.
So the next logical question would be, “why not put longer rods in all high rpm PWC modified engines” …. Simple … there are very few options to choose from.   Here is a list of various engines and their rod length to stroke ratios.  In short, the higher the ratio number…the better.
Stroke Engine Rod length Rod to stroke ratio
60.0m Kaw 400-440-550 110mm 1.833
68.0m Yam633/701/760/1100/1131 125mm 1.838
70.0m Kaw 650 all 133mm 1.900
71.0m Kaw 900 triple 133mm 1.873
71.0m Kaw 1100 triple 135mm 1.900
74.0m Kaw 750 800 all years 133mm 1.797
74.0m Sea Doo 785 140mm 1.891
74.0m Yam twin 6mm stroker 125mm 1.689
78.0m Kaw 750/800 4mm stroker 133mm 1.705
78.0m Yam GP800  GPR1176 135mm 1.730
78.0m Kaw 1200 140mm 1.794
78.2m Sea Doo 951  all 140mm 1.790
79.0m Kaw KX500  Honda CR500 145/144 1.835
Reading the Ratio Data – While the range of ratios does not look like much … a very small increase in ratio makes a very big difference.  In short, a ratio below 1.8 is not good for a high rpm engine … and those lower ratios will result in piston acceleration rates that greatly increase the risk of connecting rod failure (ie. breaking).  Of all these engines, it’s no surprise that the Kaw 650s and 1100s have the highest ratio … The Kaw 650s and 1100s both have excellent reputations for good long term crank life.  The table also clearly shows that the aftermarket Kaw and Yamaha stroker motors (that use stock rods) are very badly in need of longer connecting rods (to have good high rpm reliability).

The Overview -  For recreational 440/550 riders looking to install a more powerful (and very reliable) engine package …. A Kaw 650 has much better reliability prospects than a 750/800.  For recreational SuperJet and SXi owners looking to upgrade, the Yam 1100/1131s and Kaw 900/1100 are candidate engines that will yield excellent long term reliability.

We have prepared many engines for owners fitting the Kaw 1100 engines into standups.  We find that many of them are under propped and spinning their 1100 motors 7500-7600rpm … and that is not smart.  We know from our testing with 1100 3seaters that the best peak power is had at about 7300 rpm.  Spinning higher than that does not increase speeds…. But it does greatly increase the risk of rod failures.
For Sea Doo HX/X4 hull owners looking to upgrade, the SD 785 engine has far better reliability prospects than the 951.

I am certain that there will be many owners of various boat/engine combinations that will want to dispute the validity of the math or data shown here … and that’s fine.  After 40 years of building race engines, I have learned (the hard way) that you will never win an argument with mathematics or the laws of physics …. Never.

By Harry Klemm


There are no comments yet. Sign in to leave one!

Related products